Basic Adams Motor Theory

Tim Harwood M.A. 2001 - 2004 v. 1.03timharwood@usa.net

Comprehensive theory for the Adams motor is substantially complex. All this document attempts todo is set out a basic framework from which one can begin analysis, and nothing more. Quite simplyfor best output the mix of variables is too great, and the mechanical rotor presents too manyproblems for control, for any casual experimenter to obtain best performance. Since furthervariables beyond those discussed below exist, this document can only be considered a beginnersguide to the Adams motor. The focus is placed upon explaining how to manifest the in register fluxanomaly, rather than optimizing the anomaly itself.

Background

The first person to attempt a theory for the Adams motor was Harold Aspden. His ideas arepreserved for posterity in his patent application, reproduced in whole at the end of this document.The basic thesis Mr Aspden developed, was that the Adams motor belonged to the switchedreluctance class of motors. That is to say the magnets yawed to register, where the stator waspulsed, in effect demagnetizing the cores, so that the rotor could free wheel away from the statorzone. The over-unity effect was argued to be based upon an effect within the core, related tooperation beyond the normal flux saturation point of the material employed.

While undoubtedly clever and scientific, it must be stated these ideas were developed from theoryalone, and the motor Mr Aspden subsequently developed suffered from low speed. This isconsistent with core saturation. I am unaware of any practical engineering advantages to coresaturation. So the alternative theory I developed to explain the motor, and which I was first topropose, stated that while the rotor magnets were attracted to the stator cores, the basis of theclaimed in register anomaly, was not operation beyond the normal saturation limits of the core, butrather a generator wind on the stators, as per the cryptic 4:1 rule stated by Robert Adams (but neverillustrated in any of his diagrams).

When in register, I hypothesized the rotor magnets provided free precharge to the motor circuitry.The field of the permanent magnet reduced the net input of electrical energy required todemagnetize the stator cores. Thus it required less electrical energy to demagnetize the stator cores,than the sum of the kinetic energy we gained on approach.

This is clearly elementary physics that makes good sense. A magnet is mechanically forced into aparadoxical situation, in the sense that recession from the stator zone, requires a smaller input ofelectrical energy, than the kinetic energy gained on approach. It is not unreasonable to suppose thatin such circumstances when the input switch is closed, something anomalous, not readilydocumented in the mainstream scientific literature, may indeed happen.

The hypothesis I advanced to explain this anomaly, was that at the moment the timing switch closedand potential was extracted from the permanent magnet, a brief anomalous flux effect wasmanifested, associated with the central pole face of permanent magnets. The flux anomalyapparently coupled with the supplied stator input current, bestowing it with the noted novel andexotic qualities, input current draw reduction, above supply voltage gain, and a reversal of currentdirection back to the original source after the switch was opened.

These are general characteristics I have speculated may be associated with phase conjugation. Thatis a reversal of the time polarity of the input electron flow. The halving of current draw, possibly aresult of two time polarities being simultaneously present, doubling the effective rate of change ofmagnetic flux and thus resultant field density, one the physical electrons, the other the anomalouscoupled flux vector. The absorption of energy from the environment in place of conventionalradiation. The backwards flow of energy to the source. While not a true phase conjugate wave in thesense it traveled back in time, this hypothetical mixed coupled wave concept is my best attempt tomake sense of the phenomena.

Subsequent solid state experiments with the POD (Power On Demand) demagnetization core,suggested that while in register, the electromagnetic pulse geometry may look something like below.This brief electromagnetic 'flare' is the consequence of the mechanical yaw to register and statordemagnetization cycle, and is responsible for the noted exotic properties of the Adams motor,current draw reduction, voltage gain, and the powerful back emf surge.

Physically the POD unit consists of a ceramic ring magnet stack, typically about 2-3 inches inlength, based upon one inch diameter magnets. Two wind layers are placed over the outer side ofthe stack, each insulated by tape. The wind direction for these layers is not important. A third windlayer is then added to the central POD core, and is connected such that when pulsed, the core isdemagnetised. This is important. An end magnet in polar sync with the ring magnet stack separatedby a 1mm air gap is then added. The system is pulsed, such that the main POD core is repeatedlydemagnetised, approximating to the mechanical demagnetisation cycle of the Adams motor.

The basic idea behind the POD core, is that the two wind layers on the ring magnet stack, provideadditional potential to the demagnetisation cycle taking place on the core. Energy is thereforecontinually taken from the ring magnet stack. The ring magnet stack seeks to compensate for thisimbalance in turn, and draws energy from the end magnet. Thus the end magnet ends up manifestingthe same anomalous in register properties, as manifested mechanically by the Adams motor.

We did try simply pulsing a stator on the face of a permanent magnet, but this not not yieldexceptional results. So far as I am aware, the POD configuration is the only reliable method bywhich a permanent magnet can be placed into an exotic state, and easily studied. Other methodsinvolve placing the magnets within cores, and various complications that tend to obscure thephysics being manifested.

Implications of Theory For ConstructionThe stator zone geometry can be modeled mathematically. One is seeking to maximize the energydifference between the initial physical attraction (kinetic energy gain), and the amount of potentialmanifested in the windings by the rotor magnets (electrical energy / potential). The greater thisdifference, the more net energy is extracted from the permanent magnet, and the more exotic andabrupt the in register stator zone flux anomaly becomes.

While I have been contacted by someone who claimed to have performed such a mathematicalanalysis, he did not share it with me, other than to state the CD motor design template wasremarkably close to what his calculations gave as the ideal.

The limiting factor for the analysis is pulse width, in the sense that the current induced decays fromthe moment the switch is closed. This is also the case for the exotic switch closure anomaly, whichis a variable not presently modeled by standard equations, but which can be approximated if treatedas current induced. Increased stator windings, require a larger rotor magnet, which requires a longerpulse width to be able to free wheel away from the stator zone, upon demagnetization.

A further consideration from a commercial point of view is manufacturing cost per unit. If forexample 30% more windings deliver only 10% better output, then one might be better served toincrease the number of rotors ganged on a shaft, or the number of individual stator and rotorinteractions, rather than increasing the amount of copper wind per stator.

Furthermore, larger rotor / stator sections tend to have increased current draw, and a commercialcase could be made for such an enlargement at the expense of efficiency, if the increase in devicethroughput was shown to be a worthwhile trade off. The point is there is never going to be a perfectuniversal rotor / stator geometry, and rotor design will always be a choice of design compromises.

Modifications For Mechanically Loaded Operation

I suggested adding magnets to the end of the stator cores, such that the yaw to register becomes anactive flux path closure, instead of a passive attraction to iron. That is to say a S pole facing out onthe rotor, is attracted to a N polarity located within the stator core, coming from a permanentmagnet. Combined with uneven layouts such as 4/3 (4 rotor magnets, 3 stators), or 5/4 (5 rotormagnets, 4 stators), significant improvements in mechanical shaft drive under load, if not rotorspeed, can be obtained over the basic symmetrical dual stator unit.

Example Adams Motor Patent

The Aspden patent was not directly based upon experimental experience, and therefore despite theundoubted eloquence of the writing, failed to capture the real essence of the Adams motortechnology. I therefore wrote this simpler patent, based in part upon the language used by MrAspden, to illustrate how I think the technology could have been patented, as a matter of historicalinterest.

Abstract

A set of permanent magnets fixed on a rotor are attracted to a radially arrayed stator section. Whenthe poles of the permanent magnets are in register with individual stators, they are pulsed anddemagnetised relative to the rotor section, such as to enable the rotor magnets to leave the statorzone. By this method electrical energy is converted to kinetic energy, providing motive force to themain shaft. Back emf energy manifested by this process, may be recovered from the windings.

FIELD OF THE INVENTION

This invention relates to the field of switched reluctance motors, where electrical energy isconverted into kinetic energy, or more specifically, rotor torque. In certain configurations, thisdesign can also be adapted to provide a form of generator functionality.

BACKGROUND OF THE INVENTION

Switched reluctance motors have recently begun to find commercial popularity, as the advantagesof brush less operation and high efficiency start to become better appreciated. This invention relates

to a novelty in s.r. design that both decreases the input required to demagnetise the stator zones, aswell as providing for the recovery of input energy from the stator windings. Thus under certain loadscenarios, it has been demonstrated significant improvements in electrical efficiency can beobtained over prior art.

BRIEF DESCRIPTION OF THE INVENTION

The arrangement of the rotor and stator sections is not critical to the operation of the device, in asmuch as a variety of layouts may be used, depending upon the performance requirements of the unit.

If generator functionality is sought, then advantage may be found in symmetrical layouts, with thestators connected and pulsed in either series or parallel. Further advantage can also be found inusing multiple rotor sections. Commonly two rotor sections, but for units with a generator bias infunctionally, more may be preferred.

In the case of development for mechanical shaft torque, then asymmetrical layouts are generally tobe preferred, where the number of rotor poles and stator sections are unequal. Examples of suchlayouts would be 4 rotor poles and 3 stator sections, 5 rotor poles and 4 stator sections, and 8 rotorpoles and 7 stator sections. These examples assume more rotor poles, in that the stators representthe greater complexity and cost of construction. However there is no reason why layouts with agreater number of stators could not be developed, with the number of rotor poles and stators beingunequal the specific torque optimisation.

As regards construction of the stators, what is important is that the diameter of the stator cores isless than the diameter of the rotor magnets. This combined with an excessive number of turns on thestator cores, provides for an integrated generator functionality. That is to say when the rotor polesand stator sections are in sync, the field of the rotor magnet will extend over the stator windings,lowering the impedance of the circuitry.

This functionality provides for the improved performance of the motor, in as much as a reduction inthe amount of electrical energy required to demagnetise the stator is manifested. The magnitude ofthis effect can be correlated with the windings employed, most especially as regards the number ofturns, exact rotor / stator geometry, and resistance of the coils, where values in the range 6-10 ohmsare to be preferred.

It is a further characteristic of the motor, that back emf currents of unusual magnitude aremanifested. This is seen as a flow of energy back to the source, after the input switch is opened.There are various methods by which this energy may be collected, and some variations of the motoremploy this current to power secondary stator sections.

However, for most applications, the preferred embodiment it to route the back emf to a capacitor,from where it is distributed to a load. Typically, the load may be a bank of lead acid batteries, whichcan then be pulse charged.

ELECTRICAL MOTOR-GENERATORGB2282708

DATE OF A PUBLICATION: 12.04.1995

Applicant(s):Harold Aspden- SOUTHAMPTON, United KingdomRobert George Adams-New Zealand

Date of Filing: 30.09.1993

Application No: 9320215.8

INTeL6:HO2K 29/0823/5223/66/I HO2K 1/27

UKCL(Edition N):H2A AKC2 AKR 1 AK1O8 AK12O AK12 1 AK200 AK214R AK2 165 AK217R AK3O2B AK3O3R AK800

Documents Cited:GB 0547608 A US 5258697 A US 4972112 A US 4873463 A

Field of Search:UK CL (Edition M) H2A AKRR AKR1 AKR6 AKR9INT CL5 HO2K 23/62 29/08 29/10 29/12 53/00 57/00ONLINE DATABASES : WPI. CLAIMS

Agent and/or Address for Service:Harold Aspden, SOUTHAMPTON, United Kingdom

ABSTRACT

An electrodynamic motor-generator has a salient pole permanent magnet rotor interactingwith salient stator poies to form a machine operating on the magnetic reluctance principle.The intrinsic ferromagnetic power of the magnets provides the drive torque by bringing thepoles into register whilst current pulses demagnetize the stator poles as the polesseparate. In as much as less power is needed for stator demagnetization than is fed intothe reluctance drive by the thermodynamic system powering the ferromagnetic state, themachine operates regeneratively by virtue of stator winding interconnection with unequalnumber of rotor and stator poles. A rotor construction is disclosed (Fig 6, 7). The currentpulse may be such as to cause replusion of the rotor poles.

FIELD OF INVENTION

This invention relates to a form of electric motor which serves a generating function in thatthe machine can act regeneratively to develop output electrical power or can generatemechanical drive torque with unusually high efficiency in relation to electrical power input.

The field of invention is that of switched reluctance motors, meaning machines which have

salient poles and operate by virtue of the mutual magnetic attraction and / or repulsion asbetween magnetized poles. The invention particularly concerns a form of reluctance motorwhich incorporates permanent magnets to establish magnetic polarization.

BACKGROUND OF THE INVENTION

There have been proposals in the past for machines in which the relative motion ofmagnets can in some way develop unusually strong force actions which are said to resultin more power output than is supplied as electrical input.

By orthodox electrical engineering principles such suggestions have seemed to contradictaccepted principles of physics, but it is becoming increasingly evident that conformity withthe first law of thermodynamics allows a gain in the electromechanical power balanceprovided it is matched by a thermal cooling.

In this sense, one needs to extend the physical background of the cooling medium toinclude, not just the machine structure and the immediate ambient environment, but alsothe sub-quantum level of what is termed, in modern physics, the zero-point field. This isthe field associated with the Planck constant. Energy is constantly being exchanged asbetween that activity and coextensive matter forms but normally these energy fluctuationspreserve, on balance, an equilibrium condition so that this action passes unnoticed at thetechnology level.

Physicists are becoming more and more aware of the fact that, as with gravitation, somagnetism is a route by which we can gain access to the sea of energy that pervades thevacuum. Historically, the energy balance has been written in mathematical terms byassigning 'negative' potential to gravitation or magnetism. However, this is only adisguised way of saying that the vacuum field, suitably influenced by the gravitating massof a body in the locality or by magnetism in a ferromagnet has both the capacity and anurge to shed energy.

Now, however, there is growing awareness of the technological energy generatingpotential of this field background and interest is developing in techniques for 'pumping' thecoupling between matter and vacuum field to derive power from that hidden energysource. Such research may establish that this action will draw on the 2. 7K cosmicbackground temperature of the space medium through which the Earth travels at some400 km/s. The effect contemplated could well leave a cool vapour trail' in space as amachine delivering heat, or delivering a more useful electrical form of energy that willrevert to heat, travels with body Earth through that space.

In pure physics terms, relevant background is of recent record in the August 1993 issue ofPhysical Review E, vol. 48, pp. 1562-1565 under the title: 'Extracting energy and heatfrom the vacuum', authored by D.C. Cole and H. E. Puthoff. Though the connection is notreferenced in that paper, one of its author's presented experimental evidence on thattheme at an April 1993 conference held in Denver USA. The plasma power generatingdevice discussed at that conference was the subject of U. S. Patent No. 5,018,180, theinventor of record being K. R. Shoulders.

The invention, to be described below, operates by extracting energy from a magneticsystem in a motor and the relevant scientific background to this technology can beappreciated from the teachings of E.B. Moullin, a Cambridge Professor of ElectricalEngineering who was a President of the Institution of Electrical Engineers in U. K.

That prior art will be described below as part of the explanation of the operation of theinvention.

The invention presented here concerns specific structural design features of a machineadapted for robust operation, but these also have novelty and special merit in a functionaloperation. What is described is quite distinct from prior art proposals, one being a novelkind of motor proposed by Gareth Jones at a 1988 symposium held in Hull, Canada underthe auspices of the Planetary Association for Clean Energy. Jones suggested theadaptation of an automobile alternator which generates three-phase a. c. for rectificationand use as a power supply for the electrics in the automobile. This alternator has apermanent magnet rotor and Jones suggested that it could be used, with high efficiencygain and torque performance, by operating it as a motor with the three-phase windingcircuit excited so as to promote strong repulsion between the magnet poles and the statorpoles after the poles had come into register. However, the Jones machine is not oneexploiting the advantages of the invention to be described, because it is not strictly areluctance motor having salient poles on both stator and rotor. The stator poles in theJones machine are formed by the winding configuration in a slotted stator form, the manyslots being uniformly distributed around the inner circumference of the stator and notconstituting a pole system which lends itself to the magnetic flux actions to be describedby reference to the E.B. Moullin experiment.

The Jones machine operates by generating a rotating stator field which, in a sense,pushes the rotor poles forward rather than pulling them in the manner seen in the normalsynchronous motor. Accordingly, the Jones machine relies on the electric currentexcitation of the motor producing a field system which rotates smoothly but has a polaritypattern which is forced by the commutation control to keep behind the rotor poles inasserting a continuous repulsive drive.

Another prior art proposal which is distinguished from this invention is that of one of theapplicants, H. Aspden, namely the subject of U.K. Patent No. 2,234,863 (counterpart U.S.Patent Serial No. (4,975,608). Although this latter invention is concerned with extractingenergy from the field by the same physical process as the subject invention, the techniquefor accessing that energy is not optimum in respect of the structure or method used.Whereas in this earlier disclosure, the switching of the reluctance drive excited the polesin their approach phase, the subject invention, in one of its aspects, offers distinctadvantages by demagnetization or reversal of magnetization in the pole separation phaseof operation.

There are unexpected advantages in the implementation proposed by the subjectinvention, inasmuch as recent research has confirmed that it requires less input power toswitch off the mutual attraction across an air gap between a magnet and an electromagnetthan it does to switch it on. Usually, in electromagnetism, a reversal symmetry is expected,arising from conventional teaching of the way forward and back magnetomotive forcesgovern the resulting flux in a magnetic circuit. This will be further explained afterdescribing the scope of the invention.

BRIEF DESCRIPTION OF THE INVENTION

According to one aspect of the invention, an electrodynamic motor-generator machinecomprises a stator configured to provide a set of stator poles, a corresponding set ofmagnetizing windings mounted on the stator pole set, a rotor having two sections each ofwhich has a set of salient pole pieces, the rotor sections being axially spaced along theaxis of rotation of the rotor, rotor magnetization means disposed between the two rotorsections arranged to produce a unidirectional magnetic field which magnetically polarizesthe rotor poles, whereby the pole faces of one rotor section all have a north polarity andthe pole faces of the other rotor section all have a south polarity and electric circuitconnections between an electric current source and the stator magnetizing windingsarranged to regulate the operation of the machine by admitting current pulses for aduration determined according to the angular position of the rotor, which pulses have adirection tending to oppose the polarization induced in the stator by the rotor polarizationas stator and rotor poles separate from an in-register position, whereby the action of therotor magnetization means provides a reluctance motor drive force to bring stator androtor poles into register and the action of the stator magnetization windings opposes thecounterpart reluctance braking effect as the poles separate.

According to a feature of the invention, the circuit connecting the electric current sourceand the stator magnetizing windings is designed to deliver current pulses which are ofsufficient strength and duration to provide demagnetization of the stator poles as thestator and rotor poles separate from an in-register position.In this regard it is noted that inorder to suppress the reluctance drive torque or brake torque, depending upon whetherpoles are converging or separating, a certain amount of electrical power must be fed tothe magnetizing windings on the stator. In a sense these windings are really'demagnetizing windings' because the polarity of the circuit connections admit the pulsecurrent in the demagnetizing direction. However, it is more usual to refer to windings onmagnetic cores as 'magnetizing windings' even though they can function as primarywindings or secondary windings, the former serving the magnetization function with inputpower and the latter serving a demagnetizing function with return of power.

According to another feature of the invention, the circuit connecting the electric currentsource and the stator magnetizing windings is designed to deliver current pulses which areof sufficient strength and duration to provide a reversal of magnetic flux direction in thestator poles as the stator and rotor poles separate from an in-register position, whereby todraw on power supplied from the electric current source to provide additional forward drivetorque.

According to a further feature of this invention, the electric current source connected tostator magnetizing winding of a first stator pole comprises, at least partially, the electricalpulses induced in the stator magnetizing winding of a different second stator pole, thestator pole set configuration in relation to the rotor pole set configuration being such thatthe first stator pole is coming into register with a rotor pole as the second stator poleseparates from its in register position with a rotor pole.

This means that the magnetizing windings of two stator poles are connected so that bothserve a 'demagnetizing' function, one in resisting the magnetic action of the mutualattraction in pulling poles into register, an action which develops a current pulse outputand one in absorbing this current pulse, again by resisting the magnetic inter-pole actionto demagnetize the stator pole as its associated rotor pole separates.

In order to facilitate the function governed by this circuit 10 connection between statormagnetizing windings, a phase difference is needed and this is introduced by designingthe machine to have a different number of poles in a set of stator poles from the number

of rotor poles in each rotor section. Together with the dual rotor section feature, this hasthe additional merit of assuring a smoother torque action and reducing magnetic fluxfluctuations and leakage effects which contribute substantially to machine efficiency.

Thus, according to another feature of the invention, the stator configuration provides polepieces which are common to both rotor sections in the sense that when stator and rotorpoles are in-register the stator pole pieces constitute bridging members for magnetic fluxclosure in a magnetic circuit including that of the rotor magnetization means disposedbetween the two rotor sections.

Preferably, the number of poles in a set of stator poles and the number of rotor poles ineach section do not share a common integer factor, the number of rotor poles in one rotorsection is the same as that in the other rotor section and the number of poles in a statorset and the number of poles in a rotor section differs by one, with the pole faces Accordingto a further feature of the invention, the electric current source connected to a statormagnetizing winding of a first stator pole comprises, at least partially, the electrical pulsesinduced in the stator magnetizing winding of a different second stator pole, the stator poleset configuration in relation to the rotor pole set configuration being such that the firststator pole is coming into register with a rotor pole as being of sufficient angular width toassure that the magnetic flux produced by the rotor magnetization means can find acircuital magnetic flux closure route through the bridging path of a stator pole and throughcorresponding rotor poles for any angular position of the rotor.

It is also preferable from a design viewpoint for the stator pole faces of this invention tohave an angular width that is no greater than half the angular width of a rotor pole and forthe rotor sections to comprise circular steel laminations in which the rotor poles areformed as large teeth at the perimeter with the rotor magnetization means comprising amagnetic core structure the end faces of which abut two assemblies of such laminationsforming the two rotor sections.

According to a further feature of the invention, the rotor magnetization means comprisesat least one permanent magnet located with its polarization axis parallel with the rotor axis.The motor-generator may include an apertured metal disc that is of a non-magnetizablesubstance mounted on a rotor shaft and positioned intermediate the two rotor sections,each aperture providing location for a permanent magnet, whereby the centrifugal forcesacting on the permanent magnet as the rotor rotates are absorbed by the stresses set upin the disc. Also, the rotor may be mounted on a shaft that is of a non-magnetizablesubstance, whereby to minimize magnetic leakage from the rotor magnetizing meansthrough that shaft.

According to another aspect of the invention, an electrodynamic motor-generator machinecomprises a stator configured to provide a set of stator poles, a corresponding set ofmagnetizing windings mounted on the stator pole set, a rotor having two sections each ofwhich has a set of salient pole pieces, the rotor sections being axially spaced along theaxis of rotation of the rotor, rotor magnetization means incorporated in the rotor structureand arranged to polarize the rotor poles, whereby the pole faces of one rotor section allhave a north polarity and the pole faces of the other rotor section all have a south polarityand electric circuit connections between an electric current source and the statormagnetizing windings arranged to regulate the operation of the machine by admittingcurrent pulses for a duration determined according to the angular position of the rotor,which pulses have a direction tending to oppose the polarization induced in the stator bythe rotor polarization as stator and rotor poles separate from an in-register position,whereby the action of the rotor magnetization means provides a reluctance motor driveforce to bring stator and rotor poles into register and the action of the stator magnetizationwindings opposes the counterpart reluctance braking effect as the poles separate.

According to a feature of this latter aspect of the invention, the electric current sourceconnected to a stator magnetizing winding of a first stator pole comprises, at leastpartially, the electrical pulses induced in the stator magnetizing winding of a differentsecond stator pole, the stator pole set configuration in relation to the rotor pole setconfiguration being such that the first stator pole is coming into register with a rotor poleas the second stator pole separates from its in-register position with a rotor pole.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 presents magnetic core test data showing how the volt-amp reactance powerrequired to set up a constant magnetic flux action in an air gap, as assured by constant a.c. voltage excitation of a magnetizing winding, falls short of the associated power of thepotential implicit in the force action across that air gap.

Fig. 2 depicts the test structure to which Fig. I data applies.

Fig. 3 depicts the magnetization action at work in causing magnetic 5 flux to traverse anairgap and turn a corner in a circuit through a magnetic core.

Fig. 4 shows the configuration of a test device used to prove the operating principles ofthe invention described

.

Fig. 5 in its several illustrations depicts the progressive rotor pole to stator polerelationship as a rotor turns through a range of angular positions in a preferredembodiment of a machine according to the invention. Fig. 6 shows the form of a disc member which provides location for four permanentmagnets in the machine described.

Fig. 7 shows a cross-section of the magnetic circuit structure of a machine embodying theinvention.

Fig. 8 shows a six stator pole configuration with a seven pole rotor and depicts aschematic series connected linking of the magnetizing windings of diametrically oppositestator poles.

DETAILED DESCRIPTION OF THE INVENTION

The fact that one can extract energy from the source which powers the intrinsicferromagnetic state is not explicitly evident from existing textbooks, but it is implicit and,indeed, does become explicit once pointed out, in one textbook authored by F. B. Moullin.

His book 'The Principles of Electromagnetism' published by Clarendon Press, Oxford (3rd

Edition, 1955) describes on pages 168-174 an experiment concerned with the effect of airgaps between poles in a magnetic circuit. The data obtained are reproduced in Fig. 1,where Professor Moullin shows a curve representing a. c. current input for different airgaps, given that the voltage supplied is constant. In the same figure, Moullin presents thetheoretical current that would need to be applied to sustain the same voltage, and so therelated pole forces across the air gap, assuming (a) no flux leakage and (b) that there iscomplete equality between inductive energy input and the mechanical energy potential forthe magnetization that is established in the air gap in a quarter-cycle period at the a. c.power excitation frequency.

The data show that, even though the level of magnetic polarization is well below thesaturation value, being confined to a range that is regarded as the linear permeabilityrange in transformer design, there is a clear drop-off of current, and so the volt-ampreactive power input needed, as current increases, compared with that predicted by themechanical potential built up in the air gaps.

Unless leakage flux is excessive, here was clear evidence of anomalous energy activity.

Moullin discusses the leakage flux inferred by this experiment but points out that there isconsiderable mystery in why the effect of a small gap, which should certainly not result inmuch flux leakage in the gap region, nevertheless has an enormous effect in causing whathas to be substantial leakage in the light of the energy discrepancy.

Moullin did not contemplate that energy had been fed in from the zero-point field systemand so he left the issue with the statement that it was virtually impossible to predictleakage flux by calculation.

He was, of course, aware of magnetic domain structure and his argument was that theleakage flux problem was connected with what he termed a 'yawing' action of the flux as itpasses around the magnetic circuit. Normally, provided the level of polarization is belowthe knee of the B-H curve, which occurs at about 70% of saturation in iron cores ofgeneral crystal composition, it requires very little magnetizing field to change the magneticflux density. This is assuming that every effort is made to avoid air gaps. The actioninvolves domain wall movements so that the magnetic states of adjacent domains switchto different crystal axes of easy magnetization and this involves very little energy change.

However, if there is an air gap ahead in the flux circuit and the magnetizing winding is notsitting on that air gap, the iron core itself has to be the seat of a progressive field sourcelinking the winding and the gap. It can only serve in that sense by virtue of the lines of fluxin the domains being forced to rotate somewhat from the preferred easy axes ofmagnetization, with the help of the boundary surfaces around the whole core. This actionmeans that, forcibly, and consequential upon the existence of the air gap, the flux must becarried through the core by that 'yawing' action. It means that substantial energy is neededto force the establishment of those fields within the iron core. More important, however,from the point of view of this invention, it means that the intrinsic magnetic polarizationeffects in adjacent magnetic domains in the iron cease to be mutually parallel ororthogonal so as to stay directed along axes of easy magnetization. Then, in effect, themagnetizing action is not just that of the magnetizing winding wrapped around the core butbecomes also that of adjacent ferromagnetic polarization as the latter act in concert asvacuum-energy powered solenoids and are deflected into one another to develop theadditional forward magnetomotive forces.

The consequences of this are that the intrinsic ferromagnetic power source with itsthermodynamic ordering action contributes to doing work in building up forces across theair gap. The task, in technological terms, is then to harness that energy as the gap is

closed, as by poles coming together in a reluctance motor, and avoid returning that energyas the poles separate, this being possible if the controlling source of primarymagnetization is well removed from the pole gap and the demagnetization occurs whenthe poles are at the closest position.

This energy situation is evident in the Moullin data, because the constant a. c. voltageimplies a constant flux amplitude across the air gap if there is no flux leakage in the gapregion. A constant flux amplitude implies a constant force between the poles and so thegap width in relation to this force is a measure of the mechanical energy potential of theair gap. The reactive volt-amp power assessment over the quarter-cycle periodrepresenting the polarization demand can then be compared with the mechanical energyso made available. As already stated, this is how Moullin deduced the theoretical currentcurve. In fact, as his data show, he needed less current than the mechanical energysuggested and so he had in his experiment evidence of the vacuum energy source thatpassed unnoticed and is only now revealing itself in machines that can serve our energyneeds.

In the research leading to this patent application the Moullin experiment has beenrepeated to verify a condition where a single magnetizing winding serves three air gaps.The Moullin test configuration is shown in Fig. 2, but in repeating the experiment in theresearch leading to this invention, a search coil was mounted on the bridging member andthis was used to compare the ratio of the voltage applied to the magnetizing winding andthat induced in the search coil. The same fall-off feature in current demand was observed,and there was clear evidence of substantial excess energy in the air gap. This was inaddition to the inductive energy that necessarily had to be locked into the magnetic core tosustain the 'yawing' action of the magnetic flux already mentioned.

It is therefore emphasized that, in priming the flux 'yawing' action, energy is storedinductively in the magnetic core, even though this has been deemed to be the energy offlux leakage outside the core. The air gap energy is also induction energy. Both energiesare returned to the source winding when the system is demagnetized, given a fixed airgap. If, however, the air gap closes after or during magnetization, much of that inductiveenergy goes into the mechanical work output. Note then that the energy released asmechanical work is not just that stored in the air gap but is that stored in sustaining the'yaw'. Here, then is reason to expect an even stronger contribution to the dynamicmachine performance, one that was not embraced by the calculation of the steady-statesituation.

Given the above explanation of the energy source, the structural features which are thesubject of this invention will now be described.

The 'yawing' action is depicted in Fig. 3, which depicts how magnetic flux navigates aright-angled bend in a magnetic core upon passage through an air gap. By over-simplification it is assumed that the core has a crystal structure that has a preferred axis ofmagnetization along the broken line path. With no air gap, the current needed by amagnetizing winding has only to provide enough magnetomotive force to overcome theeffects of non-magnetic inclusions and impurities in the core substance and very highmagnetic permeabilities can apply. However, as soon as the air gap develops, this coresubstance has to find a way of setting up magnetomotive force in regions extending awayfrom the locality of the magnetizing winding. It cannot do this unless its effect is sopowerful that the magnetic flux throughout the magnetic circuit through the core substanceis everywhere deflected from alignment with a preferred easy axis of magnetization.Hence the flux vectors depicted by the arrows move out of alignment with the broken lineshown.

There is a 'knock-on' effect progressing all the way around the core from the seat of themagnetizing winding and, as already stated, this harnesses the intrinsic ferromagneticpower that, in a system with no air gap, could only be affected by magnetization above theknee of the B-H curve. Magnetic flux rotation occurs above that knee, whereas in an idealcore the magnetism develops with very high permeability over a range up to that knee,because it needs very little power to displace a magnetic domain wall sideways andpromote a 90(Degree) or a 180(Degree) flux reversal. Indeed, one can have a magneticpermeability of 10,000 below the knee and 100 above the knee, the latter reducingprogressively until the substance saturates magnetically.

In the situation depicted in Figs 2 or 3 the field strength developed by the magnetizingwindings 1 on magnetic core 2 has to be higher, the greater the air gap, in order toachieve the same amount of magnetization as measured by the voltage induced in awinding (not shown) on the bridging member 3. However, by virtue of that air gap there ispotential for harnessing energy supplied to that air gap by the intrinsic zero-point field thataccounts for the magnetic permeability being over unity and here one can contemplatevery substantial excess energy potential, give incorporation in a machine design whichdeparts from convention.

One of the applicants has built an operative test machine which is configured as depictedschematically in Fig. 4. The machine has been proved to deliver substantially moremechanical power output than is supplied as electrical input, as much as a ratio of 7:1 inone version, anc it can act regeneratively to produce electrical power.

What is shown in Fig. 4 is a simple model designed to demonstrate the principle ofoperation. It comprises a rotor in which four permanent magnets 4 are arrayed to form fourpoles. The magnets are bonded into four sectors of a non-magnetic disc 5 using a highdensity polyurethane foam filler and the composite disc is then assembled on a brassspindle 6between a split flange coupling. Not shown in the figure is the structure holdingthe spindle vertically in bearings or the star wheel commutator assembly attached to theupper shaft of the spindle.

Note that the magnets present north poles at the perimeter of the rotor disc and that thesouth poles are held together by being fimly set in the bonding material.

A series of four stator poles were formed using magnetic cores from standardelectromagnetic relays are were positioned around the rotor disc as shown. Themagnetizing windings 7 on these cores are shown to be connected in series and poweredthrough commutator contacts 8 by a d. c. power supply. Two further stator cores formedby similar electromagnetic relay components are depicted by their windings 9 in theintermediate angle positions shown and these are connected in series and connected to arectifier 10 bridged by a capacitor 11.

The rotor spindle 6 is coupled with a mechanical drive (not shown) which harnesses thetorque developed by the motor thus formed and serves as a means for measuring outputmechanical power delivered by the machine.

In operation, assuming that the rotor poles are held initially off-register with thecorresponding stator poles and the hold is then released, the strong magnetic field actionof the permanent magnets will turn the rotor to bring the stator and rotor poles intoregister. A permanent magnet has a strong attraction for soft iron and so this initialimpulse of rotation is powered by the potential energy of the magnets.

Now, with the rotor acting as a flywheel and having inertia it will have a tendency to over-shoot the in-register pole position and that will involve a reverse attraction with the result

that the rotor will oscillate until damping action brings it to rest. However, if the contacts ofthe commutating switch are closed as the poles come first into register, the magnetizingwindings 7 will receive a current pulse which, assuming the windings are connected in theright sense, tends to demagnetize the four stator cores. This means that, as the stator androtor poles separate, the reverse attraction by the magnets is eliminated. Indeed, if thedemagnetizing current pulses supplied to the windings 4 are strong enough, the statorpoles can reverse polarity and that results in a repulsion giving forward drive to theseparating rotor poles.

The net result of this action is that the rotor will continue rotating until it passes the deadcentre angular position which allows the rotor to be attracted in the forward direction bythe stator poles 90(Degree) forward of those acting originally.

The commutating switch 8 needs only to be closed for a limited period of angular travelfollowing the top dead centre in-register position of the stator and rotor poles. The powersupplied through that switch by those pulses will cause the rotor to continue rotating andhigh speeds will be achieved as the machine develops its full motor function.

Tests on such a machine have shown that more mechanical power can be delivered thanis supplied electrically by the source powering the action through the commutating switch.The reason for this is that, whereas the energy in the air gap between rotor and statorpoles which is tapped mechanically as the poles come into register is provided by theintrinsic power of the ferromagnet, a demagnetizing winding on the part of the core systemcoupled across that air gap needs very little power to eliminate the mechanical forceacting across that air gap. Imagine such a winding on the bridging member shown in Fig.2. The action of current in that winding, which sits astride the 'yawing' flux in that bridgingmember well removed from the source action of the magnetizing windings 1, is placed tobe extremely effective in resisting the magnetizing influence communicated from adistance. Hence very little power is needed to overcome the magnetic coupling transmittedacross the air gap.

Although the mutual inductance between two spaced-apart magnetizing windings has areciprocal action, regardless of which winding is primary and which is secondary, theaction in the particular machine situation being described involves the 'solenoidal'contribution represented by the 'yawing' ferromagnetic flux action. The latter is notreciprocal inasmuch as the flux 'yaw' depends on the geometry of the system. Amagnetizing winding directing flux directly across an air gap has a different influence onthe action in the ferromagnetic core from one directing flux lateral to the air gap and thereis no reciprocity in this action.

In any event, the facts of experiment do reveal that, owing to a significant discrepancy insuch mutual interaction, more mechanical power is fed into the rotor than is supplied asinput from the electrical source.

This has been further demonstrated by using the two stator windings 9 to respond in agenerator sense to the passage of the rotor poles. An electrical pulse is induced in eachwinding by the passage of a rotor pole and this is powered by the inertia of the rotor disc5. By connecting the power so generated to charge the capacitor 11 the d. c. power supplycan be augmented to enhance the efficiency even further. Indeed, the machine is able todemonstrate the excess power delivery from the ferromagnetic system by virtue ofelectrical power generation charging a battery at a greater rate than a supply battery isdischarged.

This invention is concerned with a practical embodiment of the motor-generator principlesjust described and aims, in its preferred aspect, to provide a robust and reliable machine

in which the tooth stresses in the rotor poles, which are fluctuating stressescommunicating high reluctance drive torque, are not absorbed by a ceramic permanentmagnet liable to rupture owing to its brittle composition.

Another object is to provide a structure which can be dismantled and reassembled easilyto replace the permanent magnets, but an even more important object is that ofminimizing the stray leakage flux oscillations from the powerful permanent magnets. Theirrotation in the device depicted in Fig. 4 would cause excessive eddy-current induction innearby metal, including that of the machine itself, and such effects are minimized if theflux changes are confined to paths through steel laminations and if the source flux fromthe magnets has a symmetry or near symmetry about the axis of rotation.

Thus, the ideal design with this in mind is one where the permanent magnet is a hollowcylinder located on a non-magnetic rotor shaft, but, though that structure is within thescope of this invention, the machine described will utilize several separate permanentmagnets approximating, in function, such a cylindrical configuration.

Referring to Fig. 4, it will further be noted that the magnetic flux emerging from the northpoles will have to find its way along leakage paths through air to re-enter the south poles.For periods in each cycle of machine operation the flux will be attracted through the statorcores, but the passage through air is essential and so the power of the magnets is notused to full advantage and there are those unwanted eddy-current effects.

To overcome this problem the invention provides for two separate rotor sections and thestator poles become bridging members, which with optimum design, allow the flux fromthe magnets to find a route around a magnetic circuit with minimal leakage through air asthe flux is directed through one or other pairs of air gaps where the torque action isdeveloped.

Reference is now made to Fig. 5 and the sequence of rotor positions shown. Note that thestator pole width can be significantly smaller that that of the rotor poles. Indeed, foroperation using the principles of this invention, it is advantageous for the stator to have amuch smaller pole width so as to concentrate the effective pole region. A stator pole widthof half that of the rotor is appropriate but it may be even smaller and this has thesecondary advantage of requiring smaller magnetizing windings and so saving on the lossassociated with the current circuit.

The stator has eight pole pieces formed as bridging members 12, more clearlyrepresented in Fig. 7, which shows a sectional side view through two rotor sections 13axially spaced on a rotor shaft 14. There are four permanent magnets 15 positionedbetween these rotor sections and located in apertures 16 in a disc 17 of a non-magneticsubstance of high tensile strength, the latter being shown in Fig. 6. The rotor sections areformed from disc laminations of electrical steel which has seven large teeth, the salientpoles. Magnetizing windings 18 mounted on the bridging members 12 constitute thesystem governing the action of the motor-generator being described.

The control circuitry is not described as design of such circuitry involves ordinary skillpossessed by those involved in the electrical engineering art.

It suffices, therefore, to describe the merits of the structural design configuration of thecore elements of the machine. These concern principally the magnetic action and, as canbe imagined from Fig. 7, the magnetic flux from the magnets enters the rotor laminationsby traversing the planar faces of the laminations and being deflected into the plane of thelaminations to pass through one or other of the stator pole bridging members, returning bya similar route through the other rotor.

By using eight stator poles and seven rotor poles, the latter having a pole width equal tohalf the pole pitch in an angular sense, it will be seen from Fig. 5, that there is always aflux passage across the small air gap between stator and rotor poles. However, as onepole combination is in-register the diametrically-opposed pole combinations are out-of-register.

As described by reference to Fig. 4 the operation of the machine involves allowing themagnet to pull stator and rotor poles into register and then, as they separate, pulsing thewinding on the relevant stator member to demagnetize that member. In the Fig. 4 system,all the stator magnetizing windings were pulsed together, which is not an optimum way inwhich to drive a multi-pole machine.

In the machine having the pole structure with one less rotor pole than stator poles (or anequivalent design in which there is one less stator pole than rotor poles) this pulsing actioncan be distributed in its demand on the power supply, and though this makes thecommutation switch cicuit more expensive the resulting benefit outweighs that cost.

However, there is a feature of this invention by which that problem 15 can be alleviated ifnot eliminated.

Suppose that the rotor has the position shown in Fig. 5(a) with the rotor pole denoted R1midway between stator poles S1 and S2 imagine that this is attracted towards the in-register position with stator pole S2. Upon reaching that in-register position, as shown inFig. 5 (c), suppose that the magnetizing winding of stator pole S2 is excited by a currentpulse which is sustained until the rotor reaches the Fig. 5(e) position. The combination ofthese two actions will have imparted a forward drive impulse powered by the permanentmagnet in the rotor structure and the current pulse which suppresses braking action willhave drawn a smaller amount of energy from the electrical power source that supplies it.This is the same process as was described by reference to Fig. 4.

However, now consider the events occurring in the rotor action diametrically opposite thatjust described. In the Fig 5(a) position rotor pole R4 has come fully into register with statorpole S5 and so stator pole S5 is ready to be demagnetized. However, the magneticcoupling between the rotor and stator poles is then at its strongest. Note, however, that inthat Fig. 5(a) position R5 is beginning its separation from stator pole S6and themagnetizing winding of stator pole S6 must then begin draw power to initiatedemagnetization. During that following period of pole separation the power from themagnet is pulling R1 and S2 together with much more action than is needed to generatethat current pulse needed to demagnetize S6. It follows, therefore, that, based on theresearch findings of the regenerative excitation in the test system of Fig. 4, the seriesconnection of the magnetizing windings on stators S2 and S6 will, without needing anycommutative switching, provide the regenerative power needed for machine operation.

The complementary action of the two magnetizing windings during the pole closure andpole separation allows the construction of a machine which, given that the zero-pointvacuum energy powering the ferromagnet is feeding input power, will run on that source ofenergy and thereby cool the sustaining field system.

There are various design options in implementing what has just been proposed. Muchdepends upon the intended use of the machine. If it is intended to deliver mechanicalpower output the regenerative electrical power action can all be used to power thedemagnetization with any surplus contributing to a stronger drive torque by reversing thepolarity of the stator poles during pole separation.

If the object is to generate electricity by operating in generator mode then one coulddesign a machine having additional windings on the stator for delivering electrical poweroutput. However, it seems preferable to regard the machine as a motor and maximize itsefficiency in that capacity whilst using a mechanical coupling to an alternator ofconventional design for the electrical power generation function. In the latter case it wouldstill seem preferable to use the self-excitation feature already described to reducecommutation switching problems.

The question of providing for machine start-up can be addressed by using a separatestarter motor powered from an external supply or by providing for current pulsing limitedto, say, two stator poles. Thus, for example, with the eight stator pole configuration, thecross-connected magnetizing windings could be limited to three stator pairs, with twostator magnetizing windings left free for connection to a pulsed external supply source.

If the latter feature were not required, then the stator magnetizing windings would all beconnected in pairs on a truly diametrically opposite basis. Thus Fig. 8 shows a rotor-statorconfiguration having six stator poles interacting with seven rotor poles and statormagnetizing windings linked together in pairs.

The invention, therefore, offers a wide range of implementation possibilities, which, in thelight of this disclosure will become obvious to persons skilled in the electrical engineeringart, all based, however, on the essential but simple principle that a rotor has a set of polesof common polarity which are attracted into register with a set of stator poles that aresuppressed or reversed in polarity magnetically during pole separation. The invention,however, also offers the important feature of minimizing commutation and providingfurther for a magnetic flux closure that minimizes the leakage flux and fluctuations ofleakage flux and so contributes to efficiency and high torque performance as well asdurability and reliability of a machine incorporating the invention.

It is noted that although a machine has been described which uses two rotor sections it ispossible to build a composite version of the machine having several rotor sections. In theeventuality that the invention finds use in very large motor-generator machines theproblem of providing very large magnets can be overcome by a design in which numeroussmall magnets are assembled. The structural concept described by reference to Fig. 6 inproviding locating apertures to house the magnets makes this proposal highly feasible.Furthermore, it is possible to replace the magnets by a steel cylinder and provide asolenoid as part of the stator structure and located between the rotor sections. This wouldset up an axial magnetic field magnetizing the steel cylinder and so polarizing the rotor.However, the power supplied to that solenoid would detract from the power generated andso such a machine would not be as effective as the use of permanent magnets such asare now available. Nevertheless, should one see significant progress in the developmentof warm superconductor materials, it may become feasible to harness the self-generatingmotor-generator features of the invention, with its self cooling properties, by operating thedevice in an enclosure at low temperatures and replacing the magnets by asuperconductive statorsupported solenoid.

CLAIMS

1. An electrodynamic motor-generator machine comprising a stator configured to providea set of stator poles, a corresponding set of magnetizing windings mounted on the statorpole set, a rotor having two sections each of which has a set of salient pole pieces, therotor sections being axially spaced along the axis of rotation of the rotor, rotormagnetization means disposed between the two rotor sections arranged to produce a

unidirectional magnetic field which magnetically polarizes the rotor poles, whereby thepole faces of one rotor section all have a north polarity and the pole faces of the otherrotor section all have a south polarity and electric circuit connections between an electriccurrent source and the stator magnetizing windings arranged to regulate the operation ofthe machine by admitting current pulses for a duration determined according to theangular position of the rotor, which pulses have a direction tending to oppose thepolarization induced in the stator by the rotor polarization as stator and rotor polesseparate from an in-register position, whereby the action of the rotor magnetization meansprovides a reluctance motor drive force to bring stator and rotor poles into register and theaction of the stator magnetization windings opposes the counterpart reluctance brakingeffect as the poles separate.

2. A motor-generator according to claim 1, wherein the circuit connecting the electriccurrent source and the stator magnetizing windings is designed to deliver current pulseswhich are of sufficient strength and duration to provide demagnetization of the stator polesas the stator and rotor poles separate from an in-register position.

3. A motor-generator according to claim 1, wherein the circuit connecting the electriccurrent source and the stator magnetizing windings is designed to deliver current pulseswhich are of sufficient strength and duration to provide a reversal of magnetic fluxdirection in the stator poles as the stator and rotor poles separate from an in-registerposition, whereby to draw on power supplied from the electric current source to provideadditional forward drive torque.

4. A motor-generator according to claim 1, wherein the electric current source connectedto a stator magnetizing winding of a first stator pole comprises, at least partially, theelectrical pulses induced in the stator magnetizing winding of a different second statorpole, the stator pole set configuration in relation to the rotor pole set configuration beingsuch that the first stator pole is coming into register with a rotor pole as the second statorpole separates from its in-register position with a rotor pole.

5. A motor-generator according to claim 1, wherein the number of poles in a set of statorpoles is different from the number of rotor poles in each rotor section.

6. A motor-generator according to claim I, wherein the stator configuration provides polepieces which are common to both rotor sections in the sense that when stator and rotorpoles are in-register the stator pole pieces constitute bridging members for magnetic fluxclosure in a magnetic circuit including that of the rotor magnetization means disposedbetween the two rotor sections.

7. A motor-generator according to claim 6, wherein the number of poles in a set of statorpoles and the number of rotor poles in each section do not share a common integer factorand the number of rotor poles in one rotor section is the same as that in the other rotorsection.

8. A motor-generator according to claim 7, wherein the number of poles in a stator set andthe number of poles in a rotor section differs by one and the pole faces are of sufficientangular width to assure that the magnetic flux produced by the rotor magnetization meanscan find a circuital magnetic flux closure route through the bridging path of a stator poleand through corresponding rotor poles for any angular position of the rotor.

9. A motor-generator according to claim 8, wherein each rotor section comprises sevenpoles.

10. A motor-generator according to claim 7, wherein there are N rotor poles in each rotorsection and each has an angular width that is 180/N degree of angle.

11. A motor-generator according to claim 7, wherein the stator pole faces have an angularwidth that is no greater than half the angular width of a rotor pole.

12. A motor-generator according to claim 1, wherein the rotor sections comprise circularsteel laminations in which the rotor poles are formed as large teeth at the perimeter, andthe rotor magnetization means comprise a magnetic core structure the end faces of whichabut two assemblies of 20 such laminations forming the two rotor sections.

13. A motor-generator according to claim 1 in which the rotor magnetization meanscomprises at least one permanent magnet located with its polarization axis parallel withthe rotor axis.

14. A motor-generator according to claim 13, wherein an apertured metal disc that is of anon-magnetizable substance is mounted on a rotor shaft and positioned intermediate thetwo rotor sections and each aperture provides location for a permanent magnet, wherebythe centrifugal forces acting on the permanent magnet as the rotor rotates are absorbedby the stresses set up in the disc.

15. A motor-generator according to claim 1, having a rotor mounted on a shaft that is of anon-magnetizable substance, whereby to minimize 5 magnetic leakage from the rotormagnetizing means.

16. An electrodynamic motor-generator machine comprising a stator configured to providea set of stator poles, a corresponding set of magnetizing windings mounted on the statorpole set, a rotor having two sections each of which has a set of salient pole pieces, therotor sections being axially spaced along the axis of rotation of the rotor, rotormagnetization means incorporated in the rotor structure and arranged to polarize the rotorpoles, whereby the pole faces of one rotor section all have a north polarity and the polefaces of the other rotor section all have a south polarity and electric circuit connectionsbetween an electric current source and the stator magnetizing windings arranged toregulate the operation of the machine by admitting current pulses for a durationdetermined according to the angular position of the rotor, which pulses have a directiontending to oppose the polarization induced in the stator by the rotor polarization as statorand rotor poles separate from an in-register position, whereby the action of the rotormagnetization means provides a reluctance motor drive force to bring stator and rotorpoles into register and the action of the stator magnetization windings opposes thecounterpart reluctance braking effect as the poles separate.

17. A motor- generator according to claim 16, wherein the electric current sourceconnected to a stator magnetizing winding of a first stator pole comprises, at leastpartially, the electrical pulses induced in the stator magnetizing winding of a differentsecond stator pole, the stator pole set configuration in relation to the rotor pole setconfiguration being such that the first stator pole is coming into register with a rotor poleas the second stator pole separates from its in-register position with a rotor pole.

AMENDMENTS TO THE CLAIMS HAVE BEEN FILED AS FOLLOWS

1. An electrodynamic motor-generator machine comprising a stator configured to provide

a set of stator poles, a corresponding set of magnetizing windings mounted on the statorpole set, a rotor having two sections each of which has a set of salient pole pieces, therotor sections being axially spaced along the axis of rotation of the rotor, rotormagnetization means disposed between the two rotor sections arranged to produce aunidirectional magnetic field which magnetically polarizes the rotor poles, whereby thepole faces of one rotor section all have a north polarity and the pole faces of the otherrotor section all have a south polarity and electric circuit connections between an electriccurrent source and the stator magnetizing windings arranged to regulate the operation ofthe machine by admitting current pulses for a duration determined according to theangular position of the rotor, which pulses have a direction tending to oppose thepolarization induced in the stator by the rotor polarization as stator and rotor polesseparate from an in-register position, whereby the action of the rotor magnetization meansprovides a reluctance motor drive force to bring stator and rotor poles into register and theaction of the stator magnetization windings opposes the counterpart reluctance brakingeffect as the poles separate, the machine being characterized in that the stator comprisesseparate ferromagnetic bridging members mounted parallel with the rotor axis, the ends ofwhich constitute stator poles and the core sect ions of which provide cross-sectiondisposed antiparallel with the unidirectional magnetic field polarization axis of the rotormagnetizing means.

2. A motor-generator according to claim 1, wherein the circuit connecting the electriccurrent source and the stator magnetizing windinga is designed to deliver current pulseswhich are of sufficient strength and duration to provide demagnetization of the stator polesas the stator and rotor poles separate from an in-register position.

3. A motor-generator according to claim 1, wherein the circuit connecting the electriccurrent source and the stator magnetizing windings is designed to deliver current pulseswhich are of sufficient strength and duration to provide a reversal of magnetic fluxdirection in the stator poles as the stator and rotor poles separate from an in- registerposition, whereby to draw on power supplied from the electric current source to provideadditional forward drive torque.

4. A motor-generator according to claim 1, wherein the electric current source connectedto a stator magnetizing winding of a first stator pole comprises, at least partially, theelectrical pulses induced in the stator magnetizing winding of a different second etatorpole, the stator pole set configuration in relation to the rotor pole set configuration beingsuch that the first stator pole is coming into register with a rotor pole as the second statorpole separates from its in-register position with a rotor pole.

5. A motor-generator according to claim 1, wherein the number of poles in a set of atatorpoles is different from the number of rotor poles in each rotor section.

6. A motor-generator according to claim 1, wherein the stator configuration provides polepieces which are common to both rotor sections in the sense that when stator and rotorpoles are in-register the stator pole pieces constitute bridging members for magnetic fluxclosure in a magnetic circuit including that of the rotor magnetization means disposedbetween the two rotor sections.

7. A motor-generator according to claim 6, wherein the number of poles in a set of statorpoles and the number of rotor poles in each section do not share a common integer factorand the number of rotor poles in one rotor section is the same as that in the other rotorsection.